
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns


data = pd.read_csv('WA_Fn-UseC_-HR-Employee-Attrition.csv')
data.head()


data.tail()


data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 1470 entries, 0 to 1469
Data columns (total 35 columns):
 #   Column                    Non-Null Count  Dtype 
---  ------                    --------------  ----- 
 0   Age                       1470 non-null   int64 
 1   Attrition                 1470 non-null   object
 2   BusinessTravel            1470 non-null   object
 3   DailyRate                 1470 non-null   int64 
 4   Department                1470 non-null   object
 5   DistanceFromHome          1470 non-null   int64 
 6   Education                 1470 non-null   int64 
 7   EducationField            1470 non-null   object
 8   EmployeeCount             1470 non-null   int64 
 9   EmployeeNumber            1470 non-null   int64 
 10  EnvironmentSatisfaction   1470 non-null   int64 
 11  Gender                    1470 non-null   object
 12  HourlyRate                1470 non-null   int64 
 13  JobInvolvement            1470 non-null   int64 
 14  JobLevel                  1470 non-null   int64 
 15  JobRole                   1470 non-null   object
 16  JobSatisfaction           1470 non-null   int64 
 17  MaritalStatus             1470 non-null   object
 18  MonthlyIncome             1470 non-null   int64 
 19  MonthlyRate               1470 non-null   int64 
 20  NumCompaniesWorked        1470 non-null   int64 
 21  Over18                    1470 non-null   object
 22  OverTime                  1470 non-null   object
 23  PercentSalaryHike         1470 non-null   int64 
 24  PerformanceRating         1470 non-null   int64 
 25  RelationshipSatisfaction  1470 non-null   int64 
 26  StandardHours             1470 non-null   int64 
 27  StockOptionLevel          1470 non-null   int64 
 28  TotalWorkingYears         1470 non-null   int64 
 29  TrainingTimesLastYear     1470 non-null   int64 
 30  WorkLifeBalance           1470 non-null   int64 
 31  YearsAtCompany            1470 non-null   int64 
 32  YearsInCurrentRole        1470 non-null   int64 
 33  YearsSinceLastPromotion   1470 non-null   int64 
 34  YearsWithCurrManager      1470 non-null   int64 
dtypes: int64(26), object(9)
memory usage: 402.1+ KB


data.describe()


data.isnull().sum()

Age                         0
Attrition                   0
BusinessTravel              0
DailyRate                   0
Department                  0
DistanceFromHome            0
Education                   0
EducationField              0
EmployeeCount               0
EmployeeNumber              0
EnvironmentSatisfaction     0
Gender                      0
HourlyRate                  0
JobInvolvement              0
JobLevel                    0
JobRole                     0
JobSatisfaction             0
MaritalStatus               0
MonthlyIncome               0
MonthlyRate                 0
NumCompaniesWorked          0
Over18                      0
OverTime                    0
PercentSalaryHike           0
PerformanceRating           0
RelationshipSatisfaction    0
StandardHours               0
StockOptionLevel            0
TotalWorkingYears           0
TrainingTimesLastYear       0
WorkLifeBalance             0
YearsAtCompany              0
YearsInCurrentRole          0
YearsSinceLastPromotion     0
YearsWithCurrManager        0
dtype: int64


cor = data.corr()

<ipython-input-57-06847dd9a2e1>:1: FutureWarning: The default value of numeric_only in DataFrame.corr is deprecated. In a future version, it will default to False. Select only valid columns or specify the value of numeric_only to silence this warning.
  cor = data.corr()


fig, ax = plt.subplots(figsize=(25,25))
sns.heatmap(cor, annot=True)

<Axes: >


sns.pairplot(data)

<seaborn.axisgrid.PairGrid at 0x7b7a7577e530>


from sklearn.preprocessing import LabelEncoder


le=LabelEncoder()


data["BusinessTravel"]=le.fit_transform(data["BusinessTravel"])


data["Department"]=le.fit_transform(data["Department"])


data["EducationField"]=le.fit_transform(data["EducationField"])


data["Gender"]=le.fit_transform(data["Gender"])


data["JobRole"]=le.fit_transform(data["JobRole"])


data["MaritalStatus"]=le.fit_transform(data["MaritalStatus"])


data["Over18"]=le.fit_transform(data["Over18"])


data["OverTime"]=le.fit_transform(data["OverTime"])


data.head()


data.tail()


X=data.drop(columns=["EmployeeNumber","EmployeeCount","StandardHours","Attrition","Over18"],axis=1)


y=data["Attrition"]


from sklearn.preprocessing import MinMaxScaler
ms=MinMaxScaler()


X_Scaled=ms.fit_transform(X)


cor=data.corr()

<ipython-input-76-410fe4458127>:1: FutureWarning: The default value of numeric_only in DataFrame.corr is deprecated. In a future version, it will default to False. Select only valid columns or specify the value of numeric_only to silence this warning.
  cor=data.corr()


fig, ax = plt.subplots(figsize=(30,30))
sns.heatmap(cor, annot=True)

<Axes: >


from sklearn.model_selection import train_test_split
x_train,x_test,y_train,y_test = train_test_split(X_Scaled,y,test_size =0.2,random_state =0)


from sklearn.linear_model import LogisticRegression
classifier = LogisticRegression(random_state=0)
classifier.fit(x_train,y_train)

LogisticRegression(random_state=0)

LogisticRegression(random_state=0)


from sklearn.metrics import accuracy_score,confusion_matrix
y_pred = classifier.predict(x_test)
cm = confusion_matrix(y_test, y_pred)
print(cm)
accuracy_score(y_test, y_pred)*100

[[242   3]
 [ 32  17]]

88.09523809523809


from sklearn.metrics import accuracy_score,confusion_matrix,classification_report,roc_auc_score,roc_curve


print(classification_report(y_test,y_pred))

              precision    recall  f1-score   support

          No       0.88      0.99      0.93       245
         Yes       0.85      0.35      0.49        49

    accuracy                           0.88       294
   macro avg       0.87      0.67      0.71       294
weighted avg       0.88      0.88      0.86       294


from sklearn.tree import DecisionTreeClassifier
dtc=DecisionTreeClassifier()


dtc.fit(x_train,y_train)

DecisionTreeClassifier()

DecisionTreeClassifier()


from sklearn.metrics import accuracy_score,confusion_matrix
y_pred = dtc.predict(x_test)
cm = confusion_matrix(y_test, y_pred)
print(cm)
accuracy_score(y_test, y_pred)*100

[[207  38]
 [ 33  16]]

75.85034013605441


from sklearn import tree
plt.figure(figsize=(25,15))
tree.plot_tree(dtc,filled=True)

[Text(0.32152859669811323, 0.9722222222222222, 'x[23] <= 0.038\ngini = 0.269\nsamples = 1176\nvalue = [988, 188]'),
 Text(0.07547169811320754, 0.9166666666666666, 'x[14] <= 0.75\ngini = 0.5\nsamples = 78\nvalue = [39, 39]'),
 Text(0.04716981132075472, 0.8611111111111112, 'x[4] <= 0.554\ngini = 0.426\nsamples = 39\nvalue = [27, 12]'),
 Text(0.031446540880503145, 0.8055555555555556, 'x[13] <= 0.167\ngini = 0.312\nsamples = 31\nvalue = [25, 6]'),
 Text(0.018867924528301886, 0.75, 'x[18] <= 0.5\ngini = 0.49\nsamples = 7\nvalue = [3, 4]'),
 Text(0.012578616352201259, 0.6944444444444444, 'x[8] <= 0.5\ngini = 0.375\nsamples = 4\nvalue = [3, 1]'),
 Text(0.006289308176100629, 0.6388888888888888, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.018867924528301886, 0.6388888888888888, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.025157232704402517, 0.6944444444444444, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.0440251572327044, 0.75, 'x[17] <= 0.056\ngini = 0.153\nsamples = 24\nvalue = [22, 2]'),
 Text(0.03773584905660377, 0.6944444444444444, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.050314465408805034, 0.6944444444444444, 'x[7] <= 0.167\ngini = 0.083\nsamples = 23\nvalue = [22, 1]'),
 Text(0.0440251572327044, 0.6388888888888888, 'x[20] <= 0.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.03773584905660377, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.050314465408805034, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.05660377358490566, 0.6388888888888888, 'gini = 0.0\nsamples = 21\nvalue = [21, 0]'),
 Text(0.06289308176100629, 0.8055555555555556, 'x[19] <= 0.679\ngini = 0.375\nsamples = 8\nvalue = [2, 6]'),
 Text(0.05660377358490566, 0.75, 'gini = 0.0\nsamples = 6\nvalue = [0, 6]'),
 Text(0.06918238993710692, 0.75, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.10377358490566038, 0.8611111111111112, 'x[9] <= 0.364\ngini = 0.426\nsamples = 39\nvalue = [12, 27]'),
 Text(0.0880503144654088, 0.8055555555555556, 'x[15] <= 0.1\ngini = 0.133\nsamples = 14\nvalue = [1, 13]'),
 Text(0.08176100628930817, 0.75, 'gini = 0.0\nsamples = 13\nvalue = [0, 13]'),
 Text(0.09433962264150944, 0.75, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.11949685534591195, 0.8055555555555556, 'x[18] <= 0.5\ngini = 0.493\nsamples = 25\nvalue = [11, 14]'),
 Text(0.1069182389937107, 0.75, 'x[2] <= 0.106\ngini = 0.484\nsamples = 17\nvalue = [10, 7]'),
 Text(0.10062893081761007, 0.6944444444444444, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.11320754716981132, 0.6944444444444444, 'x[21] <= 0.167\ngini = 0.408\nsamples = 14\nvalue = [10, 4]'),
 Text(0.10062893081761007, 0.6388888888888888, 'x[23] <= 0.013\ngini = 0.375\nsamples = 4\nvalue = [1, 3]'),
 Text(0.09433962264150944, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.1069182389937107, 0.5833333333333334, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.12578616352201258, 0.6388888888888888, 'x[3] <= 0.75\ngini = 0.18\nsamples = 10\nvalue = [9, 1]'),
 Text(0.11949685534591195, 0.5833333333333334, 'gini = 0.0\nsamples = 9\nvalue = [9, 0]'),
 Text(0.1320754716981132, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.1320754716981132, 0.75, 'x[15] <= 0.103\ngini = 0.219\nsamples = 8\nvalue = [1, 7]'),
 Text(0.12578616352201258, 0.6944444444444444, 'gini = 0.0\nsamples = 7\nvalue = [0, 7]'),
 Text(0.13836477987421383, 0.6944444444444444, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.5675854952830188, 0.9166666666666666, 'x[18] <= 0.5\ngini = 0.235\nsamples = 1098\nvalue = [949, 149]'),
 Text(0.31087853773584906, 0.8611111111111112, 'x[25] <= 0.167\ngini = 0.162\nsamples = 798\nvalue = [727, 71]'),
 Text(0.1792452830188679, 0.8055555555555556, 'x[2] <= 0.747\ngini = 0.38\nsamples = 47\nvalue = [35, 12]'),
 Text(0.17295597484276728, 0.75, 'x[10] <= 0.5\ngini = 0.463\nsamples = 33\nvalue = [21, 12]'),
 Text(0.1509433962264151, 0.6944444444444444, 'x[4] <= 0.446\ngini = 0.42\nsamples = 10\nvalue = [3, 7]'),
 Text(0.14465408805031446, 0.6388888888888888, 'gini = 0.0\nsamples = 6\nvalue = [0, 6]'),
 Text(0.15723270440251572, 0.6388888888888888, 'x[5] <= 0.125\ngini = 0.375\nsamples = 4\nvalue = [3, 1]'),
 Text(0.1509433962264151, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.16352201257861634, 0.5833333333333334, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.1949685534591195, 0.6944444444444444, 'x[24] <= 0.583\ngini = 0.34\nsamples = 23\nvalue = [18, 5]'),
 Text(0.18238993710691823, 0.6388888888888888, 'x[28] <= 0.333\ngini = 0.117\nsamples = 16\nvalue = [15, 1]'),
 Text(0.1761006289308176, 0.5833333333333334, 'gini = 0.0\nsamples = 15\nvalue = [15, 0]'),
 Text(0.18867924528301888, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.20754716981132076, 0.6388888888888888, 'x[26] <= 0.287\ngini = 0.49\nsamples = 7\nvalue = [3, 4]'),
 Text(0.20125786163522014, 0.5833333333333334, 'gini = 0.0\nsamples = 4\nvalue = [0, 4]'),
 Text(0.2138364779874214, 0.5833333333333334, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.18553459119496854, 0.75, 'gini = 0.0\nsamples = 14\nvalue = [14, 0]'),
 Text(0.4425117924528302, 0.8055555555555556, 'x[23] <= 0.975\ngini = 0.145\nsamples = 751\nvalue = [692, 59]'),
 Text(0.43622248427672955, 0.75, 'x[26] <= 0.113\ngini = 0.143\nsamples = 750\nvalue = [692, 58]'),
 Text(0.3018867924528302, 0.6944444444444444, 'x[7] <= 0.167\ngini = 0.218\nsamples = 257\nvalue = [225, 32]'),
 Text(0.2468553459119497, 0.6388888888888888, 'x[29] <= 0.147\ngini = 0.355\nsamples = 65\nvalue = [50, 15]'),
 Text(0.22641509433962265, 0.5833333333333334, 'x[29] <= 0.029\ngini = 0.303\nsamples = 59\nvalue = [48, 11]'),
 Text(0.20440251572327045, 0.5277777777777778, 'x[10] <= 0.5\ngini = 0.463\nsamples = 22\nvalue = [14, 8]'),
 Text(0.1918238993710692, 0.4722222222222222, 'x[9] <= 0.179\ngini = 0.198\nsamples = 9\nvalue = [8, 1]'),
 Text(0.18553459119496854, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.19811320754716982, 0.4166666666666667, 'gini = 0.0\nsamples = 8\nvalue = [8, 0]'),
 Text(0.2169811320754717, 0.4722222222222222, 'x[9] <= 0.4\ngini = 0.497\nsamples = 13\nvalue = [6, 7]'),
 Text(0.21069182389937108, 0.4166666666666667, 'gini = 0.0\nsamples = 4\nvalue = [4, 0]'),
 Text(0.22327044025157233, 0.4166666666666667, 'x[4] <= 0.286\ngini = 0.346\nsamples = 9\nvalue = [2, 7]'),
 Text(0.2169811320754717, 0.3611111111111111, 'x[28] <= 0.1\ngini = 0.444\nsamples = 3\nvalue = [2, 1]'),
 Text(0.21069182389937108, 0.3055555555555556, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.22327044025157233, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.22955974842767296, 0.3611111111111111, 'gini = 0.0\nsamples = 6\nvalue = [0, 6]'),
 Text(0.24842767295597484, 0.5277777777777778, 'x[13] <= 0.167\ngini = 0.149\nsamples = 37\nvalue = [34, 3]'),
 Text(0.24213836477987422, 0.4722222222222222, 'x[26] <= 0.088\ngini = 0.5\nsamples = 6\nvalue = [3, 3]'),
 Text(0.2358490566037736, 0.4166666666666667, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.24842767295597484, 0.4166666666666667, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.25471698113207547, 0.4722222222222222, 'gini = 0.0\nsamples = 31\nvalue = [31, 0]'),
 Text(0.2672955974842767, 0.5833333333333334, 'x[10] <= 0.167\ngini = 0.444\nsamples = 6\nvalue = [2, 4]'),
 Text(0.2610062893081761, 0.5277777777777778, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.27358490566037735, 0.5277777777777778, 'x[20] <= 0.5\ngini = 0.444\nsamples = 3\nvalue = [2, 1]'),
 Text(0.2672955974842767, 0.4722222222222222, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.279874213836478, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.35691823899371067, 0.6388888888888888, 'x[0] <= 0.321\ngini = 0.161\nsamples = 192\nvalue = [175, 17]'),
 Text(0.3081761006289308, 0.5833333333333334, 'x[6] <= 0.1\ngini = 0.294\nsamples = 67\nvalue = [55, 12]'),
 Text(0.3018867924528302, 0.5277777777777778, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.31446540880503143, 0.5277777777777778, 'x[25] <= 0.5\ngini = 0.26\nsamples = 65\nvalue = [55, 10]'),
 Text(0.29245283018867924, 0.4722222222222222, 'x[6] <= 0.5\ngini = 0.469\nsamples = 16\nvalue = [10, 6]'),
 Text(0.2861635220125786, 0.4166666666666667, 'gini = 0.0\nsamples = 7\nvalue = [7, 0]'),
 Text(0.29874213836477986, 0.4166666666666667, 'x[7] <= 0.833\ngini = 0.444\nsamples = 9\nvalue = [3, 6]'),
 Text(0.29245283018867924, 0.3611111111111111, 'gini = 0.0\nsamples = 5\nvalue = [0, 5]'),
 Text(0.3050314465408805, 0.3611111111111111, 'x[16] <= 0.072\ngini = 0.375\nsamples = 4\nvalue = [3, 1]'),
 Text(0.29874213836477986, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.3113207547169811, 0.3055555555555556, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.33647798742138363, 0.4722222222222222, 'x[2] <= 0.037\ngini = 0.15\nsamples = 49\nvalue = [45, 4]'),
 Text(0.330188679245283, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.34276729559748426, 0.4166666666666667, 'x[2] <= 0.938\ngini = 0.117\nsamples = 48\nvalue = [45, 3]'),
 Text(0.33647798742138363, 0.3611111111111111, 'x[5] <= 0.875\ngini = 0.081\nsamples = 47\nvalue = [45, 2]'),
 Text(0.3238993710691824, 0.3055555555555556, 'x[10] <= 0.167\ngini = 0.043\nsamples = 45\nvalue = [44, 1]'),
 Text(0.31761006289308175, 0.25, 'x[13] <= 0.5\ngini = 0.444\nsamples = 3\nvalue = [2, 1]'),
 Text(0.3113207547169811, 0.19444444444444445, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.3238993710691824, 0.19444444444444445, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.330188679245283, 0.25, 'gini = 0.0\nsamples = 42\nvalue = [42, 0]'),
 Text(0.3490566037735849, 0.3055555555555556, 'x[29] <= 0.088\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.34276729559748426, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.3553459119496855, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.3490566037735849, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.4056603773584906, 0.5833333333333334, 'x[6] <= 0.9\ngini = 0.077\nsamples = 125\nvalue = [120, 5]'),
 Text(0.39308176100628933, 0.5277777777777778, 'x[0] <= 0.393\ngini = 0.05\nsamples = 118\nvalue = [115, 3]'),
 Text(0.3867924528301887, 0.4722222222222222, 'x[2] <= 0.956\ngini = 0.185\nsamples = 29\nvalue = [26, 3]'),
 Text(0.3805031446540881, 0.4166666666666667, 'x[10] <= 0.167\ngini = 0.133\nsamples = 28\nvalue = [26, 2]'),
 Text(0.36792452830188677, 0.3611111111111111, 'x[2] <= 0.216\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.36163522012578614, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.3742138364779874, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.39308176100628933, 0.3611111111111111, 'x[29] <= 0.147\ngini = 0.074\nsamples = 26\nvalue = [25, 1]'),
 Text(0.3867924528301887, 0.3055555555555556, 'gini = 0.0\nsamples = 24\nvalue = [24, 0]'),
 Text(0.39937106918238996, 0.3055555555555556, 'x[22] <= 0.333\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.39308176100628933, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.4056603773584906, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.39308176100628933, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.39937106918238996, 0.4722222222222222, 'gini = 0.0\nsamples = 89\nvalue = [89, 0]'),
 Text(0.41823899371069184, 0.5277777777777778, 'x[9] <= 0.514\ngini = 0.408\nsamples = 7\nvalue = [5, 2]'),
 Text(0.4119496855345912, 0.4722222222222222, 'x[29] <= 0.059\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.4056603773584906, 0.4166666666666667, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.41823899371069184, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.42452830188679247, 0.4722222222222222, 'gini = 0.0\nsamples = 4\nvalue = [4, 0]'),
 Text(0.570558176100629, 0.6944444444444444, 'x[26] <= 0.787\ngini = 0.1\nsamples = 493\nvalue = [467, 26]'),
 Text(0.5373427672955975, 0.6388888888888888, 'x[13] <= 0.5\ngini = 0.094\nsamples = 486\nvalue = [462, 24]'),
 Text(0.48977987421383645, 0.5833333333333334, 'x[12] <= 0.938\ngini = 0.154\nsamples = 191\nvalue = [175, 16]'),
 Text(0.4834905660377358, 0.5277777777777778, 'x[16] <= 0.481\ngini = 0.145\nsamples = 190\nvalue = [175, 15]'),
 Text(0.4638364779874214, 0.4722222222222222, 'x[29] <= 0.794\ngini = 0.221\nsamples = 95\nvalue = [83, 12]'),
 Text(0.45754716981132076, 0.4166666666666667, 'x[16] <= 0.47\ngini = 0.207\nsamples = 94\nvalue = [83, 11]'),
 Text(0.45125786163522014, 0.3611111111111111, 'x[5] <= 0.375\ngini = 0.192\nsamples = 93\nvalue = [83, 10]'),
 Text(0.42452830188679247, 0.3055555555555556, 'x[6] <= 0.9\ngini = 0.363\nsamples = 21\nvalue = [16, 5]'),
 Text(0.41823899371069184, 0.25, 'x[15] <= 0.413\ngini = 0.266\nsamples = 19\nvalue = [16, 3]'),
 Text(0.4056603773584906, 0.19444444444444445, 'x[4] <= 0.982\ngini = 0.117\nsamples = 16\nvalue = [15, 1]'),
 Text(0.39937106918238996, 0.1388888888888889, 'gini = 0.0\nsamples = 14\nvalue = [14, 0]'),
 Text(0.4119496855345912, 0.1388888888888889, 'x[19] <= 0.071\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.4056603773584906, 0.08333333333333333, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.41823899371069184, 0.08333333333333333, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.4308176100628931, 0.19444444444444445, 'x[19] <= 0.893\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.42452830188679247, 0.1388888888888889, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.4371069182389937, 0.1388888888888889, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.4308176100628931, 0.25, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.4779874213836478, 0.3055555555555556, 'x[27] <= 0.139\ngini = 0.129\nsamples = 72\nvalue = [67, 5]'),
 Text(0.46226415094339623, 0.25, 'x[15] <= 0.244\ngini = 0.444\nsamples = 6\nvalue = [4, 2]'),
 Text(0.4559748427672956, 0.19444444444444445, 'x[17] <= 0.278\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.449685534591195, 0.1388888888888889, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.46226415094339623, 0.1388888888888889, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.46855345911949686, 0.19444444444444445, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.4937106918238994, 0.25, 'x[9] <= 0.993\ngini = 0.087\nsamples = 66\nvalue = [63, 3]'),
 Text(0.4811320754716981, 0.19444444444444445, 'x[24] <= 0.583\ngini = 0.061\nsamples = 64\nvalue = [62, 2]'),
 Text(0.4748427672955975, 0.1388888888888889, 'gini = 0.0\nsamples = 51\nvalue = [51, 0]'),
 Text(0.48742138364779874, 0.1388888888888889, 'x[12] <= 0.812\ngini = 0.26\nsamples = 13\nvalue = [11, 2]'),
 Text(0.4811320754716981, 0.08333333333333333, 'gini = 0.0\nsamples = 9\nvalue = [9, 0]'),
 Text(0.4937106918238994, 0.08333333333333333, 'x[7] <= 0.5\ngini = 0.5\nsamples = 4\nvalue = [2, 2]'),
 Text(0.48742138364779874, 0.027777777777777776, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.5, 0.027777777777777776, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.5062893081761006, 0.19444444444444445, 'x[26] <= 0.263\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.5, 0.1388888888888889, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5125786163522013, 0.1388888888888889, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.4638364779874214, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.470125786163522, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5031446540880503, 0.4722222222222222, 'x[17] <= 0.5\ngini = 0.061\nsamples = 95\nvalue = [92, 3]'),
 Text(0.4968553459119497, 0.4166666666666667, 'gini = 0.0\nsamples = 76\nvalue = [76, 0]'),
 Text(0.5094339622641509, 0.4166666666666667, 'x[29] <= 0.088\ngini = 0.266\nsamples = 19\nvalue = [16, 3]'),
 Text(0.4968553459119497, 0.3611111111111111, 'x[2] <= 0.547\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.49056603773584906, 0.3055555555555556, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.5031446540880503, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.5220125786163522, 0.3611111111111111, 'x[15] <= 0.108\ngini = 0.117\nsamples = 16\nvalue = [15, 1]'),
 Text(0.5157232704402516, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5283018867924528, 0.3055555555555556, 'gini = 0.0\nsamples = 15\nvalue = [15, 0]'),
 Text(0.4960691823899371, 0.5277777777777778, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5849056603773585, 0.5833333333333334, 'x[19] <= 0.036\ngini = 0.053\nsamples = 295\nvalue = [287, 8]'),
 Text(0.5628930817610063, 0.5277777777777778, 'x[28] <= 0.7\ngini = 0.159\nsamples = 46\nvalue = [42, 4]'),
 Text(0.5566037735849056, 0.4722222222222222, 'x[9] <= 0.071\ngini = 0.124\nsamples = 45\nvalue = [42, 3]'),
 Text(0.5408805031446541, 0.4166666666666667, 'x[5] <= 0.5\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.5345911949685535, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.5471698113207547, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5723270440251572, 0.4166666666666667, 'x[23] <= 0.688\ngini = 0.089\nsamples = 43\nvalue = [41, 2]'),
 Text(0.559748427672956, 0.3611111111111111, 'x[12] <= 0.062\ngini = 0.048\nsamples = 41\nvalue = [40, 1]'),
 Text(0.5534591194968553, 0.3055555555555556, 'x[2] <= 0.487\ngini = 0.375\nsamples = 4\nvalue = [3, 1]'),
 Text(0.5471698113207547, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.559748427672956, 0.25, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.5660377358490566, 0.3055555555555556, 'gini = 0.0\nsamples = 37\nvalue = [37, 0]'),
 Text(0.5849056603773585, 0.3611111111111111, 'x[17] <= 0.611\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.5786163522012578, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.5911949685534591, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.5691823899371069, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6069182389937107, 0.5277777777777778, 'x[15] <= 0.056\ngini = 0.032\nsamples = 249\nvalue = [245, 4]'),
 Text(0.5911949685534591, 0.4722222222222222, 'x[14] <= 0.75\ngini = 0.32\nsamples = 5\nvalue = [4, 1]'),
 Text(0.5849056603773585, 0.4166666666666667, 'gini = 0.0\nsamples = 4\nvalue = [4, 0]'),
 Text(0.5974842767295597, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6226415094339622, 0.4722222222222222, 'x[2] <= 0.015\ngini = 0.024\nsamples = 244\nvalue = [241, 3]'),
 Text(0.610062893081761, 0.4166666666666667, 'x[2] <= 0.013\ngini = 0.278\nsamples = 6\nvalue = [5, 1]'),
 Text(0.6037735849056604, 0.3611111111111111, 'gini = 0.0\nsamples = 5\nvalue = [5, 0]'),
 Text(0.6163522012578616, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6352201257861635, 0.4166666666666667, 'x[21] <= 0.167\ngini = 0.017\nsamples = 238\nvalue = [236, 2]'),
 Text(0.6289308176100629, 0.3611111111111111, 'x[25] <= 0.833\ngini = 0.073\nsamples = 53\nvalue = [51, 2]'),
 Text(0.6163522012578616, 0.3055555555555556, 'x[29] <= 0.088\ngini = 0.041\nsamples = 48\nvalue = [47, 1]'),
 Text(0.610062893081761, 0.25, 'x[16] <= 0.824\ngini = 0.245\nsamples = 7\nvalue = [6, 1]'),
 Text(0.6037735849056604, 0.19444444444444445, 'gini = 0.0\nsamples = 6\nvalue = [6, 0]'),
 Text(0.6163522012578616, 0.19444444444444445, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6226415094339622, 0.25, 'gini = 0.0\nsamples = 41\nvalue = [41, 0]'),
 Text(0.6415094339622641, 0.3055555555555556, 'x[28] <= 0.367\ngini = 0.32\nsamples = 5\nvalue = [4, 1]'),
 Text(0.6352201257861635, 0.25, 'gini = 0.0\nsamples = 4\nvalue = [4, 0]'),
 Text(0.6477987421383647, 0.25, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6415094339622641, 0.3611111111111111, 'gini = 0.0\nsamples = 185\nvalue = [185, 0]'),
 Text(0.6037735849056604, 0.6388888888888888, 'x[8] <= 0.5\ngini = 0.408\nsamples = 7\nvalue = [5, 2]'),
 Text(0.5974842767295597, 0.5833333333333334, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.610062893081761, 0.5833333333333334, 'gini = 0.0\nsamples = 5\nvalue = [5, 0]'),
 Text(0.4488011006289308, 0.75, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.8242924528301887, 0.8611111111111112, 'x[15] <= 0.157\ngini = 0.385\nsamples = 300\nvalue = [222, 78]'),
 Text(0.7382075471698113, 0.8055555555555556, 'x[22] <= 0.167\ngini = 0.5\nsamples = 96\nvalue = [49, 47]'),
 Text(0.7044025157232704, 0.75, 'x[4] <= 0.161\ngini = 0.459\nsamples = 42\nvalue = [15, 27]'),
 Text(0.6792452830188679, 0.6944444444444444, 'x[16] <= 0.41\ngini = 0.499\nsamples = 23\nvalue = [12, 11]'),
 Text(0.660377358490566, 0.6388888888888888, 'x[15] <= 0.061\ngini = 0.426\nsamples = 13\nvalue = [4, 9]'),
 Text(0.6540880503144654, 0.5833333333333334, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.6666666666666666, 0.5833333333333334, 'x[9] <= 0.993\ngini = 0.298\nsamples = 11\nvalue = [2, 9]'),
 Text(0.660377358490566, 0.5277777777777778, 'x[0] <= 0.214\ngini = 0.18\nsamples = 10\nvalue = [1, 9]'),
 Text(0.6540880503144654, 0.4722222222222222, 'x[2] <= 0.406\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.6477987421383647, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.660377358490566, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.6666666666666666, 0.4722222222222222, 'gini = 0.0\nsamples = 8\nvalue = [0, 8]'),
 Text(0.6729559748427673, 0.5277777777777778, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.6981132075471698, 0.6388888888888888, 'x[7] <= 0.167\ngini = 0.32\nsamples = 10\nvalue = [8, 2]'),
 Text(0.6918238993710691, 0.5833333333333334, 'x[2] <= 0.544\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.6855345911949685, 0.5277777777777778, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.6981132075471698, 0.5277777777777778, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.7044025157232704, 0.5833333333333334, 'gini = 0.0\nsamples = 7\nvalue = [7, 0]'),
 Text(0.7295597484276729, 0.6944444444444444, 'x[11] <= 0.125\ngini = 0.266\nsamples = 19\nvalue = [3, 16]'),
 Text(0.7232704402515723, 0.6388888888888888, 'x[9] <= 0.2\ngini = 0.198\nsamples = 18\nvalue = [2, 16]'),
 Text(0.7169811320754716, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.7295597484276729, 0.5833333333333334, 'x[28] <= 0.433\ngini = 0.111\nsamples = 17\nvalue = [1, 16]'),
 Text(0.7232704402515723, 0.5277777777777778, 'gini = 0.0\nsamples = 15\nvalue = [0, 15]'),
 Text(0.7358490566037735, 0.5277777777777778, 'x[29] <= 0.235\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.7295597484276729, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.7421383647798742, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.7358490566037735, 0.6388888888888888, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.7720125786163522, 0.75, 'x[0] <= 0.202\ngini = 0.466\nsamples = 54\nvalue = [34, 20]'),
 Text(0.7547169811320755, 0.6944444444444444, 'x[0] <= 0.107\ngini = 0.245\nsamples = 7\nvalue = [1, 6]'),
 Text(0.7484276729559748, 0.6388888888888888, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.7610062893081762, 0.6388888888888888, 'gini = 0.0\nsamples = 6\nvalue = [0, 6]'),
 Text(0.789308176100629, 0.6944444444444444, 'x[2] <= 0.622\ngini = 0.418\nsamples = 47\nvalue = [33, 14]'),
 Text(0.7735849056603774, 0.6388888888888888, 'x[2] <= 0.145\ngini = 0.482\nsamples = 32\nvalue = [19, 13]'),
 Text(0.7610062893081762, 0.5833333333333334, 'x[26] <= 0.237\ngini = 0.18\nsamples = 10\nvalue = [9, 1]'),
 Text(0.7547169811320755, 0.5277777777777778, 'gini = 0.0\nsamples = 9\nvalue = [9, 0]'),
 Text(0.7672955974842768, 0.5277777777777778, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.7861635220125787, 0.5833333333333334, 'x[16] <= 0.87\ngini = 0.496\nsamples = 22\nvalue = [10, 12]'),
 Text(0.779874213836478, 0.5277777777777778, 'x[25] <= 0.833\ngini = 0.465\nsamples = 19\nvalue = [7, 12]'),
 Text(0.7735849056603774, 0.4722222222222222, 'x[17] <= 0.167\ngini = 0.415\nsamples = 17\nvalue = [5, 12]'),
 Text(0.7610062893081762, 0.4166666666666667, 'x[19] <= 0.321\ngini = 0.49\nsamples = 7\nvalue = [4, 3]'),
 Text(0.7547169811320755, 0.3611111111111111, 'gini = 0.0\nsamples = 4\nvalue = [4, 0]'),
 Text(0.7672955974842768, 0.3611111111111111, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.7861635220125787, 0.4166666666666667, 'x[12] <= 0.188\ngini = 0.18\nsamples = 10\nvalue = [1, 9]'),
 Text(0.779874213836478, 0.3611111111111111, 'x[19] <= 0.393\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.7735849056603774, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.7861635220125787, 0.3055555555555556, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.7924528301886793, 0.3611111111111111, 'gini = 0.0\nsamples = 8\nvalue = [0, 8]'),
 Text(0.7861635220125787, 0.4722222222222222, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.7924528301886793, 0.5277777777777778, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.8050314465408805, 0.6388888888888888, 'x[9] <= 0.064\ngini = 0.124\nsamples = 15\nvalue = [14, 1]'),
 Text(0.7987421383647799, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.8113207547169812, 0.5833333333333334, 'gini = 0.0\nsamples = 14\nvalue = [14, 0]'),
 Text(0.910377358490566, 0.8055555555555556, 'x[14] <= 0.75\ngini = 0.258\nsamples = 204\nvalue = [173, 31]'),
 Text(0.8553459119496856, 0.75, 'x[15] <= 0.992\ngini = 0.138\nsamples = 147\nvalue = [136, 11]'),
 Text(0.8490566037735849, 0.6944444444444444, 'x[4] <= 0.482\ngini = 0.128\nsamples = 146\nvalue = [136, 10]'),
 Text(0.8301886792452831, 0.6388888888888888, 'x[26] <= 0.063\ngini = 0.038\nsamples = 104\nvalue = [102, 2]'),
 Text(0.8238993710691824, 0.5833333333333334, 'x[9] <= 0.193\ngini = 0.32\nsamples = 10\nvalue = [8, 2]'),
 Text(0.8176100628930818, 0.5277777777777778, 'x[15] <= 0.599\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.8113207547169812, 0.4722222222222222, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.8238993710691824, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.8301886792452831, 0.5277777777777778, 'gini = 0.0\nsamples = 7\nvalue = [7, 0]'),
 Text(0.8364779874213837, 0.5833333333333334, 'gini = 0.0\nsamples = 94\nvalue = [94, 0]'),
 Text(0.8679245283018868, 0.6388888888888888, 'x[7] <= 0.167\ngini = 0.308\nsamples = 42\nvalue = [34, 8]'),
 Text(0.8490566037735849, 0.5833333333333334, 'x[2] <= 0.736\ngini = 0.375\nsamples = 4\nvalue = [1, 3]'),
 Text(0.8427672955974843, 0.5277777777777778, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.8553459119496856, 0.5277777777777778, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.8867924528301887, 0.5833333333333334, 'x[0] <= 0.393\ngini = 0.229\nsamples = 38\nvalue = [33, 5]'),
 Text(0.8679245283018868, 0.5277777777777778, 'x[1] <= 0.25\ngini = 0.5\nsamples = 6\nvalue = [3, 3]'),
 Text(0.8616352201257862, 0.4722222222222222, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.8742138364779874, 0.4722222222222222, 'x[5] <= 0.625\ngini = 0.375\nsamples = 4\nvalue = [1, 3]'),
 Text(0.8679245283018868, 0.4166666666666667, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.8805031446540881, 0.4166666666666667, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.9056603773584906, 0.5277777777777778, 'x[24] <= 0.917\ngini = 0.117\nsamples = 32\nvalue = [30, 2]'),
 Text(0.89937106918239, 0.4722222222222222, 'x[12] <= 0.812\ngini = 0.062\nsamples = 31\nvalue = [30, 1]'),
 Text(0.8930817610062893, 0.4166666666666667, 'gini = 0.0\nsamples = 28\nvalue = [28, 0]'),
 Text(0.9056603773584906, 0.4166666666666667, 'x[0] <= 0.69\ngini = 0.444\nsamples = 3\nvalue = [2, 1]'),
 Text(0.89937106918239, 0.3611111111111111, 'gini = 0.0\nsamples = 2\nvalue = [2, 0]'),
 Text(0.9119496855345912, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.9119496855345912, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.8616352201257862, 0.6944444444444444, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.9654088050314465, 0.75, 'x[12] <= 0.812\ngini = 0.456\nsamples = 57\nvalue = [37, 20]'),
 Text(0.9433962264150944, 0.6944444444444444, 'x[28] <= 0.4\ngini = 0.238\nsamples = 29\nvalue = [25, 4]'),
 Text(0.9308176100628931, 0.6388888888888888, 'x[9] <= 0.964\ngini = 0.142\nsamples = 26\nvalue = [24, 2]'),
 Text(0.9245283018867925, 0.5833333333333334, 'x[20] <= 0.5\ngini = 0.077\nsamples = 25\nvalue = [24, 1]'),
 Text(0.9182389937106918, 0.5277777777777778, 'gini = 0.0\nsamples = 23\nvalue = [23, 0]'),
 Text(0.9308176100628931, 0.5277777777777778, 'x[12] <= 0.375\ngini = 0.5\nsamples = 2\nvalue = [1, 1]'),
 Text(0.9245283018867925, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.9371069182389937, 0.4722222222222222, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.9371069182389937, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.9559748427672956, 0.6388888888888888, 'x[28] <= 0.933\ngini = 0.444\nsamples = 3\nvalue = [1, 2]'),
 Text(0.949685534591195, 0.5833333333333334, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.9622641509433962, 0.5833333333333334, 'gini = 0.0\nsamples = 1\nvalue = [1, 0]'),
 Text(0.9874213836477987, 0.6944444444444444, 'x[28] <= 0.1\ngini = 0.49\nsamples = 28\nvalue = [12, 16]'),
 Text(0.9811320754716981, 0.6388888888888888, 'x[4] <= 0.804\ngini = 0.48\nsamples = 20\nvalue = [12, 8]'),
 Text(0.9748427672955975, 0.5833333333333334, 'x[26] <= 0.013\ngini = 0.415\nsamples = 17\nvalue = [12, 5]'),
 Text(0.9685534591194969, 0.5277777777777778, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.9811320754716981, 0.5277777777777778, 'x[21] <= 0.5\ngini = 0.32\nsamples = 15\nvalue = [12, 3]'),
 Text(0.9748427672955975, 0.4722222222222222, 'x[7] <= 0.167\ngini = 0.5\nsamples = 6\nvalue = [3, 3]'),
 Text(0.9685534591194969, 0.4166666666666667, 'gini = 0.0\nsamples = 2\nvalue = [0, 2]'),
 Text(0.9811320754716981, 0.4166666666666667, 'x[24] <= 0.083\ngini = 0.375\nsamples = 4\nvalue = [3, 1]'),
 Text(0.9748427672955975, 0.3611111111111111, 'gini = 0.0\nsamples = 1\nvalue = [0, 1]'),
 Text(0.9874213836477987, 0.3611111111111111, 'gini = 0.0\nsamples = 3\nvalue = [3, 0]'),
 Text(0.9874213836477987, 0.4722222222222222, 'gini = 0.0\nsamples = 9\nvalue = [9, 0]'),
 Text(0.9874213836477987, 0.5833333333333334, 'gini = 0.0\nsamples = 3\nvalue = [0, 3]'),
 Text(0.9937106918238994, 0.6388888888888888, 'gini = 0.0\nsamples = 8\nvalue = [0, 8]')]


from sklearn.model_selection import GridSearchCV
parameter={
 'criterion':['gini','entropy'],
  'splitter':['best','random'],
  'max_depth':[1,2,3,4,5,6,7,8,9,10],
  'max_features':['sqrt']

}


grid_search=GridSearchCV(estimator=dtc,param_grid=parameter,cv=5,scoring="accuracy")


grid_search.fit(x_train,y_train)

GridSearchCV(cv=5, estimator=DecisionTreeClassifier(),
             param_grid={'criterion': ['gini', 'entropy'],
                         'max_depth': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
                         'max_features': ['sqrt'],
                         'splitter': ['best', 'random']},
             scoring='accuracy')

GridSearchCV(cv=5, estimator=DecisionTreeClassifier(),
             param_grid={'criterion': ['gini', 'entropy'],
                         'max_depth': [1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
                         'max_features': ['sqrt'],
                         'splitter': ['best', 'random']},
             scoring='accuracy')

DecisionTreeClassifier()

DecisionTreeClassifier()


grid_search.best_params_

{'criterion': 'entropy',
 'max_depth': 5,
 'max_features': 'auto',
 'splitter': 'random'}


dtc_cv=DecisionTreeClassifier(criterion= 'entropy',max_depth= 4,max_features= 'sqrt',splitter= 'best')
dtc_cv.fit(x_train,y_train)

DecisionTreeClassifier(criterion='entropy', max_depth=4, max_features='sqrt')

DecisionTreeClassifier(criterion='entropy', max_depth=4, max_features='sqrt')


print(classification_report(y_test,y_pred))

              precision    recall  f1-score   support

          No       0.86      0.84      0.85       245
         Yes       0.30      0.33      0.31        49

    accuracy                           0.76       294
   macro avg       0.58      0.59      0.58       294
weighted avg       0.77      0.76      0.76       294


from sklearn.ensemble import RandomForestClassifier
classifier = RandomForestClassifier(n_estimators = 1000, criterion = 'entropy', random_state = 0)
classifier.fit(x_train, y_train)

RandomForestClassifier(criterion='entropy', n_estimators=1000, random_state=0)

RandomForestClassifier(criterion='entropy', n_estimators=1000, random_state=0)


from sklearn.metrics import confusion_matrix, accuracy_score
y_pred = classifier.predict(x_test)
cm = confusion_matrix(y_test, y_pred)
print(cm)
accuracy_score(y_test, y_pred)

[[243   2]
 [ 41   8]]

0.8537414965986394


from sklearn.ensemble import RandomForestClassifier


rfc=RandomForestClassifier()


forest_params = [{'max_depth': list(range(10, 15)),'max_features': ['sqrt', 'log2', 1, 0.5, None]}]


rfc_cv=GridSearchCV(rfc,param_grid=forest_params,cv=10,scoring="accuracy")


rfc_cv.fit(x_train,y_train)

GridSearchCV(cv=10, estimator=RandomForestClassifier(),
             param_grid=[{'max_depth': [10, 11, 12, 13, 14],
                          'max_features': ['sqrt', 'log2', 1, 0.5, None]}],
             scoring='accuracy')

GridSearchCV(cv=10, estimator=RandomForestClassifier(),
             param_grid=[{'max_depth': [10, 11, 12, 13, 14],
                          'max_features': ['sqrt', 'log2', 1, 0.5, None]}],
             scoring='accuracy')

RandomForestClassifier()

RandomForestClassifier()


print(classification_report(y_test,y_pred))

              precision    recall  f1-score   support

          No       0.86      0.99      0.92       245
         Yes       0.80      0.16      0.27        49

    accuracy                           0.85       294
   macro avg       0.83      0.58      0.59       294
weighted avg       0.85      0.85      0.81       294

	Age	DailyRate	DistanceFromHome	Education	EmployeeCount	EmployeeNumber	EnvironmentSatisfaction	HourlyRate	JobInvolvement	JobLevel	...	RelationshipSatisfaction	StandardHours	StockOptionLevel	TotalWorkingYears	TrainingTimesLastYear	WorkLifeBalance	YearsAtCompany	YearsInCurrentRole	YearsSinceLastPromotion	YearsWithCurrManager
count	1470.000000	1470.000000	1470.000000	1470.000000	1470.0	1470.000000	1470.000000	1470.000000	1470.000000	1470.000000	...	1470.000000	1470.0	1470.000000	1470.000000	1470.000000	1470.000000	1470.000000	1470.000000	1470.000000	1470.000000
mean	36.923810	802.485714	9.192517	2.912925	1.0	1024.865306	2.721769	65.891156	2.729932	2.063946	...	2.712245	80.0	0.793878	11.279592	2.799320	2.761224	7.008163	4.229252	2.187755	4.123129
std	9.135373	403.509100	8.106864	1.024165	0.0	602.024335	1.093082	20.329428	0.711561	1.106940	...	1.081209	0.0	0.852077	7.780782	1.289271	0.706476	6.126525	3.623137	3.222430	3.568136
min	18.000000	102.000000	1.000000	1.000000	1.0	1.000000	1.000000	30.000000	1.000000	1.000000	...	1.000000	80.0	0.000000	0.000000	0.000000	1.000000	0.000000	0.000000	0.000000	0.000000
25%	30.000000	465.000000	2.000000	2.000000	1.0	491.250000	2.000000	48.000000	2.000000	1.000000	...	2.000000	80.0	0.000000	6.000000	2.000000	2.000000	3.000000	2.000000	0.000000	2.000000
50%	36.000000	802.000000	7.000000	3.000000	1.0	1020.500000	3.000000	66.000000	3.000000	2.000000	...	3.000000	80.0	1.000000	10.000000	3.000000	3.000000	5.000000	3.000000	1.000000	3.000000
75%	43.000000	1157.000000	14.000000	4.000000	1.0	1555.750000	4.000000	83.750000	3.000000	3.000000	...	4.000000	80.0	1.000000	15.000000	3.000000	3.000000	9.000000	7.000000	3.000000	7.000000
max	60.000000	1499.000000	29.000000	5.000000	1.0	2068.000000	4.000000	100.000000	4.000000	5.000000	...	4.000000	80.0	3.000000	40.000000	6.000000	4.000000	40.000000	18.000000	15.000000	17.000000

	Age	Attrition	BusinessTravel	DailyRate	Department	DistanceFromHome	Education	EducationField	EmployeeCount	EmployeeNumber	...	RelationshipSatisfaction	StandardHours	StockOptionLevel	TotalWorkingYears	TrainingTimesLastYear	WorkLifeBalance	YearsAtCompany	YearsInCurrentRole	YearsSinceLastPromotion	YearsWithCurrManager
0	41	Yes	Travel_Rarely	1102	Sales	1	2	Life Sciences	1	1	...	1	80	0	8	0	1	6	4	0	5
1	49	No	Travel_Frequently	279	Research & Development	8	1	Life Sciences	1	2	...	4	80	1	10	3	3	10	7	1	7
2	37	Yes	Travel_Rarely	1373	Research & Development	2	2	Other	1	4	...	2	80	0	7	3	3	0	0	0	0
3	33	No	Travel_Frequently	1392	Research & Development	3	4	Life Sciences	1	5	...	3	80	0	8	3	3	8	7	3	0
4	27	No	Travel_Rarely	591	Research & Development	2	1	Medical	1	7	...	4	80	1	6	3	3	2	2	2	2

	Age	Attrition	BusinessTravel	DailyRate	Department	DistanceFromHome	Education	EducationField	EmployeeCount	EmployeeNumber	...	RelationshipSatisfaction	StandardHours	StockOptionLevel	TotalWorkingYears	TrainingTimesLastYear	WorkLifeBalance	YearsAtCompany	YearsInCurrentRole	YearsSinceLastPromotion	YearsWithCurrManager
1465	36	No	Travel_Frequently	884	Research & Development	23	2	Medical	1	2061	...	3	80	1	17	3	3	5	2	0	3
1466	39	No	Travel_Rarely	613	Research & Development	6	1	Medical	1	2062	...	1	80	1	9	5	3	7	7	1	7
1467	27	No	Travel_Rarely	155	Research & Development	4	3	Life Sciences	1	2064	...	2	80	1	6	0	3	6	2	0	3
1468	49	No	Travel_Frequently	1023	Sales	2	3	Medical	1	2065	...	4	80	0	17	3	2	9	6	0	8
1469	34	No	Travel_Rarely	628	Research & Development	8	3	Medical	1	2068	...	1	80	0	6	3	4	4	3	1	2